论文部分内容阅读
利用开源计算机视觉库OpenCV作为图像处理工具,研究复杂环境下差分抠像技术的改进.通过RGB值补偿解决家用摄像头自动白平衡产生的颜色变化问题;消除差分抠像中对象产生的阴影;对通道优化并提取目标通道,经阈值化得到的二值图像作为遮罩以提取目标对象.实验表明,此方法可有效提取复杂环境下的前景目标,基于此种技术探讨对老年用户娱乐系统模型开发的技术支持和拓展.“,”Improving the technology of difference matting in complex scenes by using OpenCV(Open Source Computer Vision Library) as image processing tool is studied.Repair the RGB values to solve the problem of color changing which is caused by camera automatic white balance;eliminate the shadow produced by objects during difference matting;extract the target channel through optimize channels,and get the binary image as mask after threshold operation to obtain the target object.It is showed that the method can effectively extract the foreground objects in complex scenes in the experiments.At last,on the support and expansion of developing the modal of elderly entertainment system based on such technology is also discussed.